Open AccessResearch on Password Cracking Technology Based on Improved Transformer
Author(s) -
He Shen,
Jun Fu,
Cancan Chen,
Zhihui Guo
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1631/1/012161
Subject(s) - password , password strength , one time password , computer science , s/key , password cracking , cognitive password , password policy , computer security
Password plays a vital role in identity authentication. However, password security is facing great challenges. In this paper, we research the password cracking technology based on artificial intelligence, aiming to study the probability of password cracking in common password setting methods, and provide references for the setting of password. First of all, we collected a large amount of user’s personal information and passwords, and analysed the correlation between the personal information and passwords. And then, we implemented a password guessing model based on improved Transformer in which information weights were introduced into the data pre-processing and the modified beam search algorithm was used in the model to quickly search the top ranked output results. The percentage of password cracked was 68.63%, and the average guess time was 51.99 seconds. The experiment result shows that artificial intelligence brings great challenges to user password security, and this paper puts forward suggestions on user setting passwords.